Marsh vegetation affect denitrification by influencing the microbial community structure, the supply of labile organic carbon, and by introducing oxygen in the sediments via their roots which ultimately results in higher nitrate availability. Other factors such as the elevation of the marsh platform, which controls the inundation cycle and sediment redox conditions, and nutrient inputs to marshes also exert a strong control on denitrification. We examined the impact of different vegetation types on rates of denitrification by making measurements in a mixed marsh where Spartina alterniflora and Juncus roemerianus co-occur and are located at the same elevation with respect to mean sea level. We determined the microbial community structure by 16S rRNA gene sequencing on the MiSeq platform and measured potential denitrification rates in surficial and deeper sediments underlying the two vegetation types. We also measured porewater nutrients and sulfide concentrations in J. roemerianus and S. alterniflora dominated patches. We hypothesized that rates of denitrification would be higher in J. roemerianus patches where sediments tend to be less reducing.
Results/Conclusions
The 16S rRNA gene sequencing resulted in 9.8 million iTag sequences, analysis of which revealed that at the class level Anaerolineae (Chloroflexi), Deltaprotebacteria and Gammaproteobacteria had the highest average relative abundances in J. roemerianus and S. alterniflora sediment samples and at both depth intervals. In both plant types Gammaproteobacteria were most abundant, followed by Anaerolineae and Deltaproteobacteria in the surficial sediments. Differences in microbial community structure with the different plant types were, however, observed when evaluating Operational Taxonomic Units. At this taxonomic level ordination analysis revealed that samples clustered by plant type, and within each plant type, by depth interval. Additionally, the microbial community structure differed temporally, particularly in the near-surface depth interval, with samples clustering by the time of year sampled. Porewater sulfide concentrations were significantly lower in J. roemerianus sediments. Potential denitrification rates differed between plant species types across all dates and depths (ANOVA, p<0.001) and were nearly 3X higher in J. roemerianus compared to S. alterniflora patches. Potential denitrification in surficial sediments decreased during the summer for both 2015 and 2016, with the lowest rates occurring in June 2015 and July and August 2016. Similarly, potential denitrification rates in deeper sediments were lowest in July and August 2016, but they were only significantly lower than 2015 rates. Altogether, these results suggest that the higher rates of denitrification in sediments dominated by J. romerianus are mediated by the influence of the vegetation, as opposed to extrinsic factors, on the redox conditions and the microbial community structure.